TWI576670B - Network architecture for lithography machine cluster - Google Patents

Description

Network architecture for lithography etching machine clustering

The present invention is directed to a cluster substrate processing system including a plurality of lithography etch elements, each of which is configured for independent exposure of a substrate in accordance with pattern data, particularly with respect to the substrate processing system for use in the population. Network architecture.

The demand for high-accuracy and high-reliability structures in the semiconductor industry has not diminished. In lithography systems, this demand leads to extremely high demands on the quantity and speed of wafer production. One of the ways to achieve this goal is to aggregate a plurality of lithography etch elements controlled by the network architecture. The network architecture itself can thus be controlled by an automated host or by a user. Its existence combines control and data services into a single network and uses advanced service technologies to prioritize the delivery of specific data to provide a network structure for the timely delivery of critical control data, but it has proven to be a disadvantage in this environment.

The present invention provides an alternative solution for a network architecture for a group of lithographically etched elements.

It is necessary to integrate multiple lithography machines into a group, in which control and data services are included in this cluster network architecture. In one feature, the present invention provides a cluster substrate processing system including a plurality of lithography etch elements, each lithographic etch element being configured for substrate independence based on pattern data exposure. Each lithography etch element includes a complex lithography etching subsystem, a control network, and a data network configured for use between the complex lithography etching subsystem and the at least one component control unit Controlling information communication, the component control unit is configured to transmit an instruction to the complex lithography etch subsystem and the complex lithography etch subsystem is configured to transmit a response to the component control unit, the data network configured to be used Recording information communication of the plurality of lithography etching subsystems to at least one data network hub, the complex lithography etching subsystem configured to transmit data recording information to the data network hub, the data network hub being configured to Used to receive and store data record information. The system further includes a cluster front end as an interface to an operator or a master system, the cluster front end configured to transmit control information to the at least one component control unit to control the lithography etch subsystem The operation of one or more wafer exposures, and the cluster front end is further configured to receive at least a portion of the data record information received by the data hub.

The network architecture divides control and data functions into different networks. The control function requires two-way data flow and requires predictable information transmission through the control network. Data collection and management functions generally require one-way flow, moving from the lithography subsystem to the data hub and forward to the cluster interface in the data network, and can involve extremely large amounts of data. The combination of these control and data services into a single network and the use of advanced service technologies to prioritize the delivery of specific data to provide timely delivery of critical control data has proven to be inadequate in this environment. The present invention proposes an alternative solution based on some of the design features described below.

The control network forms a control network path between the component control unit and the lithography etching subsystem, and the data network forms a data network path between the data network hub and the lithography etching subsystem, and the control network The path and the data path can contain physically distinct media. The control network can have its own wiring and switches for transmitting data within the control network, which is not used in the data network. Similarly, a data network can have its own wiring and switches for transmitting data in a data network that is not used in the control network. This separation of components between the two networks prevents or reduces interference between the two networks such that high load or congestion in one network does not affect the other.

Each lithography etch subsystem can have a first connection to a control network, the first connection configured to receive an instruction from the component control unit and transmit a response to the component control unit through the control network, and each micro The shadow etching subsystem can also have a different second connection to the data network, and the second connection is configured to transmit data recording information to the data network hub through the data network. Each connection to the control and data network of each system can be used to reduce interference between simultaneous transmission and/or reception of data from the control network and transmission and/or reception of data from the data network.

The component control unit can also be configured to communicate to the data network hub via instructions from the control network to the lithography etch subsystem and related data received from a lithography etch subsystem. This connection enables the data hub to store secondary system commands and response information as well as data logging data from the secondary system. This allows two types of data to be stored in a data storage system without the need for a data network hub. Control communication on the network.

During operation of one of the lithography etch elements, the lithography etch subsystem in the lithography etch element can be configured to continuously transmit data record information to the data network hub through the data network. The secondary systems can be configured to transmit data logging information during operation without the need for command initiation from the data network hub. The data record information may include data identifying the secondary system transmitting the data record information, and measurement data obtained from the secondary system sensor and time stamp data indicating the time of the measurement.

The system can further include a data path configured for transmission of pattern data to one or more lithography etching subsystems that form a transmission path different from the control network and the data network. The data path can include a pattern processing system and a pattern stream system. The pattern data is used by the lithography etch element for the control and modulation of the beamlets to expose each of the substrates. Pattern data can be supplied from the data path to a beamlet switching or modulation subsystem to enable switching or modulation of the beamlets during exposure of a substrate. Pattern data can be transmitted to the beamlet switching or modulation subsystem without being transmitted over the control network or data network. Such pattern data typically contains a very large amount of data for each area on the substrate to be exposed and provides a data path different from the control and data network to avoid network due to the transmission of such data. congestion.

The data hub can store the information needed to initialize the lithography etch subsystem. The initialization information for a lithography etch subsystem is transferred to the lithography subsystem during the beginning of the lithography subsystem. The initialization information may include a basic operating system soft for the processor in the secondary system The default or specific settings and parameters used by the subsystems, and/or application software running on the secondary system processor to cause the secondary systems to perform their functions. The data hub can act as a boot node for these secondary systems. The data hub can also store initialization information indicating which initialization information is scheduled to be transmitted to which subsystem, so that it can store different information for different subsystems. The initial information store via the data hub provides an efficient way to provide a single repository of information for multiple subsystems and can be used to avoid requiring each subsystem to incorporate a large non-volatile memory. Store initialization information for this system. Software updates, upgrades, and/or fixes for initialization information can be downloaded to the data hub and stored there for subsequent downloads to the secondary system. This design enables efficient modification of the initialization information without disturbing the secondary system during operation and without the need to transmit over the control or data network during such modifications.

At least one of the lithography etch subsystems can be configured to transmit information to the component control unit at the start of the lithography etch subsystem, the information indicating the presence of a lithography etch subsystem and the lithography etch times on the control network One of the systems identifies the identity and indicates one or more instructions that the lithography etch subsystem can execute. The component control unit can be grouped to have no information about the identity of the secondary system connected to the control network and the instructions that each system can execute before the secondary system itself provides relevant information. This design makes it easier to modify or upgrade one of the lithographic etch elements by modifying or upgrading individual subsystems connected to the control network without modifying or upgrading the component control unit. For example, it can upgrade the software of a secondary system to enable the secondary system to execute a new or modified instruction, And you can write a new process that requires the new or modified instruction to be executed by the system. After the secondary system starts, it will report back to the component control unit indicating that it can execute the new or modified command, and the component control hub can then schedule a process job for the process program to be above the secondary system. Execute this new or modified instruction.

The control network can be configured for communication of service information between the component control unit and the lithography etch subsystem, and the cluster front end can be configured to dispense for execution on one or more lithography etch elements Service information for service functions. In this way, it is possible to perform service actions from the same interface used to control the function, and the transmission of service information does not require additional network facilities.

The system can be configured to provide a two-clock signal to the lithography etching subsystem, wherein one clock signal has a clock frequency that is at least 100 times higher than the other clock frequency, and wherein the lithography etch subsystem is configured to use The clock signals synchronize their actions. The two-clock signal provides a synchronized clock with different timing accuracy to the secondary system. For example, the accuracy of one clock signal can be milliseconds, while the accuracy of another clock signal is nanoseconds.

The lithography sub-system may not provide any interface to the operator or the host system, but rather provide such interfaces to the subsystem via only the cluster front end. The operator or master system requires only a single interface to connect to the cluster front end. It does not require an interface to multiple different subsystems to simplify the design and enable it to be modified by sub-systems without the need to modify the interface of the connection operator or the master system. A user can design their own interface to the clustered substrate processing system by fabricating a front-end interface without having to fully understand the internal architecture and protocols of the control network and data network.

In another feature, the invention comprises a method for data communication in a cluster substrate processing system comprising a plurality of lithography etch elements, each lithographic etch element being configured for pattern data The substrate is independently exposed, and each lithography etch element comprises a control network, a component control unit and a plurality of lithography etching subsystems, a data network, a data network hub and the complex lithography etching Sub-systems, as well as a group of poly front ends. The method includes transmitting control information from the component control unit to one or more of the plurality of lithography etching subsystems through the control network, and transmitting a response from the plurality of lithography etching subsystems to the component control through the control network And transmitting, by the plurality of lithography etching subsystems, data recording information to the data network hub, receiving and storing the data recording information in the data network hub, and transmitting control information from the cluster front end to the machine control unit Controlling the exposure of the plurality of lithography etching subsystems to one or more substrates, and transmitting at least a portion of the data recording information stored by the data hub to the cluster front end.

The control information transmitted to the lithography etching subsystem and the response from the lithography etching subsystem may be transmitted to the component control unit only through the control network, and the data recording information from the lithography etching subsystem may be only The data network is transmitted to the data network hub. The method can further include transmitting information from the component control unit to the data hub for instructions relating to the transfer to the lithography subsystem and the response received from a lithography subsystem. Transferring the data record information from a lithography etch sub-system to the data network hub may be performed continuously during an instruction being executed by the lithography subsystem.

The method can further include transmitting the pattern data to the one or more lithography etching subsystems via a data path, the data path forming a transmission path different from the control network and the data network. The method can also include storing information needed to initialize the lithography etch subsystem in the data hub, and transmitting the initialization information to a lithography etch subsystem during the beginning of the lithography subsystem. The method can also include transmitting information from one of the lithography etching subsystems to the component control unit at the beginning of the lithography etching system, the information indicating the presence of a lithography etching subsystem and the lithography etching on the control network. One of the secondary systems identifies the identity and indicates one or more instructions that the lithography etch subsystem can perform.

Specific embodiments of the invention are described below with reference to the drawings and are merely exemplary.

1 is a schematic illustration of one embodiment of a lithography etching system 1 having a control and data interface in accordance with the present invention. The schematic shows a hierarchical configuration with three interfaces, a population of polymeric interfaces 3, a clustering component interface 5, and a lithographic etching subsystem interface 7. The configuration illustrated in FIG. 1 has a lithography etching system cluster including a lithography etch element 10 including a plurality of lithography etching subsystems 16. The lithography etching system can include a plurality of lithography etch elements 10, as shown, for example, in the embodiment of FIG.

The clustering interface 3 includes an interface for communication between a lithography etch cluster front end 6 and one or more master systems 2 and/or between the cluster front end 6 and one or more operator consoles 4.

The clustering component interface 5 includes an interface for communication between the cluster front end 6 and a network of lithographic etch elements, the lithography etch element network component control unit 12 and/or a data network hub 14. The component control unit 12 can be coupled to a data network hub 14 via a link 106, wherein the communication is preferably unidirectionally directed from the component control unit 12 to the data network hub 14.

The lithography etch sub-system interface 7 includes an interface between the component control unit 12 and the lithography etch sub-system 16 and between the data network hub 14 and the lithography etch sub-system 16. Secondary system 16 is coupled to component control unit 12 via control network 120, and secondary system 16 is coupled to data network hub 14 via data network 140.

2 is a schematic diagram of an embodiment of a lithography etching system 1 in which a lithography etching system cluster comprises a plurality of lithography etch elements 10, each of which includes a plurality of lithography etching subsystems 16. Each component can include a component control unit 12 and a data network hub 14 that interconnects the component's lithography etch subsystem 16. Each component can be used as a separate lithography etch component that can operate independently to expose the wafer using a lithography process. In this embodiment, the plurality of lithography etch elements 10 are each connected to a single front end 6, and the front end 6 is connected to one or more main control systems 2 and/or operator interfaces that function for the entire group. 4.

The embodiment of Figures 1 and 2 is preferably designed to assist in the efficient control of a group of lithographically etched elements. Each lithography etch element 10 preferably has only a network interface to the control network 120 and the data network 140. An exception to this design rule is that the data path 20 directly connects the pattern streamer 19 to a subsystem that is responsible for modulating or switching the charged particle beam. Pattern data is initially in the map Prepared in the file processing system 18 and transmitted to the pattern streamer 19 for data conversion and streaming to the subsystem. This design focuses on the extremely large amount of pattern data that is transmitted to the relevant subsystem. Pattern data is typically streamed to the associated subsystem in a bit-map format because the amount of data is too large for local-side storage at the secondary system.

The operator interface and the interface to the higher level host management and automation computer are not comprised of individual lithographic etched elements, but are located at the cluster front end 6. This enables the development of such interfaces to be performed for clustering without the knowledge of the interface agreement of the lithography etch element 10 and without additional requirements or interference with communication over the control network 120 and the data network 140.

Figure 3 shows a simplified top view of a lithography etching system cluster 100. In this embodiment, the cluster comprises a group of ten lithographic etch elements 10 arranged in parallel to include two courses of five elements. Directly adjacent to the cluster 100, the underlying space is reserved as the service area 23. Each lithographic etch element includes an electro-optical cavity contained in its own vacuum processing chamber, with one of the vacuum processing chambers facing a substrate delivery system 22 and a service area 23. The substrate delivery system 22 receives the substrate from a substrate supply system 24 and provides it to the lithography etch element 10 for processing, and receives the processed substrate from the lithography etch element 10 and provides it to the substrate supply system 24 for transfer To other systems in the fab.

In the case of a charged particle lithography apparatus, the vacuum processing chamber preferably seals the charged particle source and assembly for generating a plurality of charged particle beamlets, and switches or modulates the beamlets of the beamlets. a variable system, a projector system for projecting the beamlets onto a predetermined patterned substrate, And a movable substrate platform. The vacuum processing chamber preferably includes a load lock system for transferring the substrate from and to the vacuum processing chamber from the substrate delivery system 22, and also includes an access door toward the service area 23 that can be opened for operation. Service to electro-optical chambers.

Each lithography etch element 10 operates independently to receive and process the wafer. Each lithography etch component includes its own computer processing system for processing data and manipulating the components and subsystems of the lithographic etch component. Each lithography etch sub-system 16 preferably has its own computer processor and memory system to perform the instructions for guiding the operation of the sub-system, collecting data generated by the operation of the sub-system, and communicating control and data networks. . Control component unit 12 and data network hub 14 preferably each include their own computer processor and memory system to perform their functions. The control unit 12 of the lithography etch element 10, the data network hub 14 and the computer processor and memory system in the subsystem 16 can be disposed at a vacuum processing chamber that is in close proximity to the lithographic etch element, for example, In the cabinet above the vacuum processing chamber, in the base below the vacuum processing chamber, or in a part of each position.

The side-by-side layout of the lithographic etched elements of the cluster allows a system to have a limited footprint, while the computer processor and memory system placed directly above or below the vacuum processing chamber also reduces the footprint. The floor space within the fab is extremely valuable, so efficient use of the floor space of the fab is important.

Figure 4 shows a simplified schematic of one of the electro-optical cavities of a charged particulate lithography etch element. For example, such a lithography etching system is described in U.S. Patent Nos. 6,897,458, 6,958,804, 7, 019, 908, 7, 084, 414, and 7, 129, 502, U.S. Patent Publication No. 2007/0064213 And the copending U.S. Patent Application Serial Nos. 61/031,573, 61/031,594, 61/045, 243, 61/055, 839, 61/058, 596, and 61/101, 682, all assigned to the owner of the present invention. Therefore, they are included in this specification in the form of their overall reference.

In the embodiment illustrated in FIG. 4, the lithography etch element cavity includes an electron source 110 that produces an expanded electron beam 130 that is collimated by a collimator lens system 113. The collimated electron beam impinges on an aperture array 114a that blocks a portion of the beam to produce a plurality of sub-beams 134 that pass through a focused sub-beam concentrator lens array 116. The sub-beams are directed onto a second array of apertures 114b that produce a plurality of beamlets 133 from each of the sub-beams 134. This system produces an extremely large number of small beams 133, preferably about 10,000 to 1,000,000 beamlets.

A beamlet blanker array 117 includes a plurality of blanking electrodes that deflect selected selected beamlets. The undeflected beamlets arrive at the beam stop array 118 and pass through a corresponding aperture, while the deflected beamlets miss the corresponding aperture and are blocked by the beam array. Thus, the beamlet blanker array 117 and the beam stop array 118 operate in concert to switch individual beamlets. The undeflected beamlets pass through the beam stop array 118 and pass through a beam deflector array 119 which deflects the beamlets to scan the beamlets across the surface of the target or substrate 121. Thereafter, the beamlets pass through the projection lens array 120 and are projected onto the substrate 121, which is placed over a movable platform for carrying the substrate. For lithography applications, the substrate typically contains a wafer with a charged particle sensing layer or photoresist layer.

The lithography etch element cavity operates in a vacuum environment. It requires a vacuum state to remove particles that may be ionized by the charged particle beam to be attracted to the source, may be freed and deposited onto the machine components, and may disperse the charged particle beam. Usually it requires a vacuum of at least 10 ^-6 bar. To maintain the vacuum environment, a charged particle lithography system is placed in a vacuum processing chamber. Preferably, all of the primary components of the lithographic etched component are housed in a common vacuum processing chamber, including a source of charged particles, a projector system for projecting a small beam onto the substrate, and a movable platform.

Secondary system

In the embodiment of Figures 1 and 2, each lithography etch element 10 forms an independently operated lithography etch element for exposing the wafer. Each lithography etch element 10 includes a plurality of subsystems 16 for performing the functions required for the components. Each secondary system performs a specific function. A typical subsystem 16 includes, for example, a wafer loading subsystem (WLS), a wafer positioning subsystem (WPS), and an illumination optical module subsystem (ILO) for generating electron beamlets. a pattern stream sub-system (PSS) for streaming beam switching data to the lithography etch element, a beam switching subsystem (BSS) for switching the electron beamlet to on or off, A projection optical module subsystem (POS), a beam measurement subsystem (BMS), a metrology system (MES), and the like for projecting a beam onto a wafer. Each lithography etch element 10 is preferably configured to receive wafers, clamp each wafer to a chuck, and prepare a clamped wafer for exposure, exposure, and clamping. The wafer, and the self-clamp release the exposed wafer and submit the exposed wafer to remove it from the component.

Each subsystem 16 can include one or more modules 17 dedicated to a particular sub-function and preferably designed as a replaceable component. Module 17 can be packaged With an actuator 19 and a sensor 21, the actuator 19 is capable of executing an instruction, and the sensor 21 is capable of detecting actions and results during, during, and/or after execution of an instruction.

Each subsystem 16 operates independently and includes a memory for storing instructions and a computer processor for executing instructions. The memory and processor can be implemented in the form of a plug-in user (PIC) 15 in each subsystem. Suitable implementations of the secondary system may include, for example, a personal computer that executes a Linux operating system. The secondary systems may include a hard disk or non-volatile memory to store their operating system such that each time the system is booted from the hard disk or memory. The features described above and other features described below facilitate a design in which each subsystem is an autonomous unit that can be designed, constructed, and tested in the form of a single unit, regardless of other subsystems. Plus the limit. For example, each subsystem can be designed to have sufficient memory and processing capacity to properly perform the functions of the subsystem during its operation, regardless of the memory and processing capacity requirements of other subsystems. This is particularly advantageous during system development and upgrades as these requirements are constantly changing. The disadvantage of this design is that the total memory and processing capacity required is increased, and too many of these components must be implemented in each subsystem. However, these shortcomings seem to be inconspicuous compared to design simplifications that make development faster and upgrade easier.

The secondary system 16 is designed to receive instructions through the control network 120 and execute the instructions in a manner independent of other secondary systems, and to report the execution results of the instructions and any execution data generated by the outgoings as required.

The secondary system 16 can be designed as an autonomous unit but designed to be self-centered The disk or memory is booted, for example, on a data hub. This reduces the reliability issues and costs of individual hard disk or non-volatile memory in each sub-system, and allows for a simpler software upgrade of the secondary system by updating the boot image of the secondary system at a central location.

Communication protocol

The secondary system 16 is coupled to a component control unit 12 via a control network, which is also referred to as a secondary subsystem control or SUSC. The component control unit 12 includes a memory and a computer processor to control the operation of the secondary system for the lithography etch component 10.

The communication 102 between the cluster front end 6 and the SUSC 12 is designed to perform PP to SUSC 12 transmission. It can use a protocol based on Java Description Object Representation (JSON). JSON is a textual open standard designed for the exchange of human readable data to represent simple data structures and arrays, and is suitable for arranging and transmitting structured data over a network connection. Preferred embodiments for use in the agreement are limited to users at the cluster, and are not limited to users outside the clean room environment. Preferably, the agreement provides an instruction for the establishment of the PJ, transmitting the PP file and any associated parameters to instruct SUSC to generate a PJ based on the PP. Other more instructions can include Abort and Cancel instructions. The communication from SUSC 12 to the cluster front end 6 can include response messages, progress reports, and error and alert messages.

The communication 101 between the SUSC 12 and the lithography etch subsystem 16 on the control network 120 is preferably tightly controlled using only component control unit protocols to ensure quasi real-time performance on the network.

Communication 105 between SUSD 14 and cluster front end 6 Acquisition of PJ results, job tracking, and data logging of SUSD 14. For such a communication link, it is possible to use Hypertext Transfer Protocol (HTTP).

The communication 103 between the lithography etch subsystem 16 and the SUSD 14 is designed for unidirectional collection of data for the subsystem 16. It can use a variety of protocols to transfer this material, such as syslog, HDF5, UDP, and other protocols.

It can use the User Data Telegraph Protocol (UDP) to transfer large amounts of data to transfer data under the enormous recurring burden of avoiding handshakes, error checking and corrections. Data can be considered to be received on the fly due to its extremely low recurring transmission burden.

The hierarchical data format HDF5 can be used for transmission and storage of high frequency data. HDF5 is ideal for storing and organizing large amounts of data, but is usually not used in a UDP environment. It can also use other data formats, such as CSV or TCP, especially for low-level (small) data.

Process program

The operation of the lithography etch element 10 is controlled using a process program (PP) 11, which includes a series of actions to be performed. The component control unit 12 is loaded with a PP and is scheduled and executed via a master control system 2 or an operator via a console 4 request. The PP 11 can take the form of a receipt and is framed to form the form defined in the SEMI E40 standard.

The PP is a set of pre-set and reusable portions of instructions, settings, and parameters that determine the processing environment of the wafer and can be changed between execution or processing cycles. It can be designed by the lithography etch tool designer PP or by tools.

It can upload the PP to the lithography etching system through the user. Its use PP generates process operations (PJ). The process operation (PJ) designates the process of applying the lithography etch component 10 to a wafer or a group of wafers. The PJ defines which PP to use when processing a particular set of wafers, and may include parameters from the PP (and optionally from the user). A PJ is a system activity initiated by a user or master system.

It can not only use PP to control the processing of wafers, but also for service operations, calibration functions, lithography etch component testing, component modification, update and upgrade software. Preferably, the secondary system only performs actions pre-assigned to the PP, except for certain allowed additional items, such as automatic initialization during startup of the module or subsystem power, periodic and unconditional actions of the secondary system ( As long as it does not affect PJ execution), and respond to an unexpected shutdown, emergency or EMO trigger.

A PP is divided into multiple steps. Most steps include an instruction and specify a secondary system that is scheduled to execute the instruction. The steps may also include parameters that are intended to be used when the instruction is executed, as well as parameter limits. The PP may also contain scheduling parameters to indicate when a step is scheduled to be performed.

For execution of one of the instruction steps of the PJ, the component control unit 12 transmits the instruction designated in the PJ to the secondary system specified in the PJ correlation step. The component control unit 12 monitors the timing and receives the results from the secondary system. An example of execution of a specific instruction is illustrated in Figures 6-13 and is explained below.

Plug-in instruction concept

A component control unit (SUSC) 12 converts a PP (which represents a logic plan) into a schedule of secondary system actions. The system can be designed such that the SUSC 12 does not have which subsystem 16 is installed, which subsystem instructions exist, and A priori knowledge of the nature. In this case, such information is provided to the SUSC 12 by the secondary system 16 at the time of execution during the initial period.

Each secondary system 16 can be designed to return its presence and performance to the SUSC 12 at the time the secondary system is powered up and initialized. The secondary system establishes communication on the control network 120 and reports its identity and status to the SUSC 12, and can also report its performance, such as which instructions it can execute. The SUSC 12 may perform a check before or during the execution of the PP 11 to check that each sub-system required for the PP has returned its presence and a ready state, and may also check that each sub-system has returned its performance. The instructions required to execute the secondary system in the PP.

The self-reward of such subsystems allows the lithography system to use only regional software upgraded to the secondary system, or even to upgrade or upgrade new features without software upgrades. For example, a particular subsystem 16 can upgrade its software to enable it to execute a new instruction. SUSC 12 can load a new PP 11, which contains new instructions for the upgraded subsystem. When the upgraded secondary system 16 is powered on, it communicates with the component control unit 12 indicating its identity and status, and which instructions it can execute, including the new instructions. The SUSC 12 schedules the PP to generate a PJ containing steps indicating that the system is executing the new instruction. SUSC 12 can perform a check to confirm that the upgraded system has reported that it can execute the new instruction. This upgrade therefore only requires an upgrade of the relevant secondary system and PP, but does not require an upgrade of SUSC 12 or any other secondary system 16.

Control network

Control network 120 preferably does not use an instant messaging protocol, but is instead Designed to provide near-instant performance, providing highly repeatable communication between the subsystem and SUSC 12, such as a repeatability of approximately one millisecond.

Such quasi-instant performance on the control network 120 can be achieved by implementing all or some of the following measures.

(a) The lithography etching system schedules each PP to generate a PJ before the PJ begins execution, and determines the time at which each step of the PJ is initiated (and may also include the time to complete each step). The start time (and completion time) scheduled for each step and the entire PJ can be completely determined before the PJ is initiated, and the time can be reported back to the cluster front end.

(b) It is possible to design PP (and PJ) to have no conditional steps or actions and no retrying actions. The PP/PJ steps describe a series of actions, although the actions within a lithographic etched element can be performed in parallel, such as instructions that are executed in parallel on different subsystems. It may program a conditional step or action into a PP by defining two (or more) alternative steps to be performed, wherein the alternative steps are arranged in parallel paths. The alternative steps are scheduled into parallel paths among the PJs, each of which is assigned the same execution time, so that the execution time of the PJ does not generally follow which of the alternative paths is selected in the secondary system. Change on execution.

The execution time of the steps of the PJ does not depend on the result of a previous step or a parallel step, but is performed in accordance with a predetermined schedule. If a step is not completed correctly or is not completed within the scheduled time, it will not be executed again (ie no retry), but the next scheduled step will be executed.

(c) After the schedule, each step will be performed by the SUSC 12 at its predetermined scheduled time point, preferably within about 1 millisecond of timing accuracy. At the scheduled time, the SUSC 12 will perform the relevant steps and transmit any instructions of this step to the associated subsystem 16 for execution by the subsystem. The scheduling timing in SUSC 12 does not depend on feedback from sub-system 16 regarding the successful completion or even failure of the previous step instruction execution. As long as a step defines an execution time length, the SUSC 12 waits for the length of time to exhaust before entering the next step of the PJ. If the secondary system returns the result of an instruction execution before the time length expires, the SUSC 12 still waits for the length of time to enter the next step of the PJ. If the secondary system fails to return an error message, the SUSC 12 waits for the length of time to expire, after which it proceeds to the next step. If the secondary system fails to execute an instruction correctly, the operator or the master system will determine what corrective action is required.

Any timing variation of the step execution is considered in the schedule so that the scheduled step execution time is greater than the maximum expected execution time. Time scheduling is usually defined in the PP. The scheduling time can be "just in time" before execution begins, or at some point before execution, for example, when a step is added to an execution queue.

(d) The instructions transmitted and executed by subsystems 16 specify a fixed maximum time for the execution of the instructions. If this time is exceeded, the operator or master system will determine what corrective action is required.

(e) when a PJ step is executed and the PJ step has instructions for the secondary system to execute and associated data transmitted to the associated subsystem, it may transmit the data to the secondary system before the instruction itself is transmitted to the secondary system. . The material is transmitted in the form of one or more messages such that any message is limited in length. The transfer of data is sufficiently advanced to ensure that it has been received by the secondary system, in other words, between the transfer of data and the transfer of instructions. The time is much longer than the expected transmission time, and the secondary system can respond to confirm receipt of the data. This allows larger messages containing data to be transmitted in advance when time is not tight, and messages containing only one of the instructions that are much smaller (without relevant information) can be transmitted at the scheduled time. This measure spreads the load on the network and avoids congestion when transmitting instructions to reduce transmission time and increase the timely and reliable transmission of instructions.

(f) Similarly, when a secondary system completes the execution of an instruction, it can send a short message to SUSC 12 indicating that the instruction has been completed, but retains any information generated by the execution of the instruction, pending that SUSC 12 later take. When the SUSC 12 receives an indication that the instruction is complete, it can then transmit a request to the secondary system to retrieve the generated data. This distributes the load on the network and avoids congestion when transmitting instruction completion indications to reduce transmission time and increase timely and reliable transmission, and can be used for data retrieval after a short time.

(g) The message exchange between the secondary system 16 and the SUSC 12 contains a reply message to confirm receipt of the command from the SUSC 12. This reply message does not carry other information except that the secondary system 16 has received the relevant information from the SUSC 12 and is still in the normal operation. This allows the SUSC 12 to detect that the secondary system 16 has failed to respond as expected, and can report back to the cluster front end in time. In order to keep the implementation of the software simple, the subsystem 16 has no time limit design. The expected total response time for a reply message is not substantially long, such as less than 0.5 milliseconds back and forth. To detect sub-systems 16 that are unresponsive, SUSC 12 has a time-limited design that far exceeds the expected response time, such as 30 seconds after an expected response is received. The limit is large enough to undoubtedly conclude that the subsystem 16 is no longer functioning, but small enough to be as usual here in an operator 4. The operator has detected a failure before, so the operator 4 does not have to suspect what happened.

(h) The control network 120 is designed to have a bandwidth that is much greater than the transmission rate used by the subsystem 16. For example, control network 120 can have a bandwidth of 1 Gbit per second, while a secondary system is designed to transmit at a rate of 100 Mbits per second. This increases the predictability of transmission time over the network.

(i) The control network uses a simple transport protocol, such as a byte concatenation on a TCP link. Avoid using transport protocols that are complex or contain unpredictable elements.

A TCP connection consists of two byte data streams, one in each direction. This protocol divides the two streams into discrete messages, each of which contains a sequence of zero or more bytes of any data, preceded by four bytes representing the length of the sequence. This length can be encoded as an unsigned integer, such as a 32-bit integer with a network order ("big-endian"), providing a maximum message length of one (2 ³² -1) bytes. This message separation mechanism can be implemented as a "Connection Object" in the Python programming language. By using the standard multiprocessing connection package, Python itself can handle this part of the contract. In other programming environments, it will need to implement this feature in a clear and detailed form. Messages can be initiated by either party as long as they conform to the specified order.

Each first message in the message sequence can contain a control data structure encoded in JavaScript Object Notation (JSON). All SUSC-compliant JSON messages can have the same basic structure: an array of two elements, the first containing a message-type string, and the second one having a named argument for the message. dictionary, For example ["<messageType>", {"param1":<value1>,"param2":<value2>,...}].

In some message sequences, such as input and output items for transmitting interface instructions, the control/command message is followed by a number of data messages, for example, one data message per input/output item. The agreement does not make any assumptions about the encoding used in the message. The interpretation is described in the design file specifying the sub-system instructions, data and behavior, for example, "two floats of tuples." The list, encoded in pickle form, or "painted as a pressure curve for image/pgn".

This mechanism of mixing arbitrary messages with JSON control messages allows the process output to be directly generated by the plug-in software into an understandable data format such as CSV, PDF, PNG, and the like. It also means that the transmission of such objects can be done in the form of a part of the data stream without additional coding or escaping, for example by using the Linux sendfile function. This degree of freedom can also be used to explicitly agree on a coding scheme that is more suitable for a particular interface function between subsystems. However, it would be unwise for each sub-system to have its own coding mechanism. Therefore, if the sub-system has no specific requirements, it should use a preset mechanism to make the best trade-off between multiple specifications. .

(j) The secondary system is not responsible for checking the correct behavior of SUSC 12 and detecting SUSC agreement errors to simplify the design of PIC 15. The purpose of the SUSC Agreement is to perform the process steps of the process program in the process operations established by the user. It uses exceptions to convey unintended behavior to the user. In addition to the errors in the execution of the instructions, of course, errors may occur due to non-compliance with the SUSC Agreement. error. SUSC 12 is responsible for recording exceptions and can close the connection to a PIC when the PIC action does not follow the SUSC agreement (or has completely stopped responding), but PIC is not responsible for checking the correct behavior of SUSC to keep the PIC specification simple. . If a PIC encounters a breach of the agreement, it can formally respond, or it can be ignored, and it does not stipulate that the subsequent message must be understood. It is recommended to try to record a number of diagnoses, record an exception, disconnect when disconnected from a message, and then reconnect. This is a recommendation, not a rule, because it cannot guarantee or expect the PIC to be in a usable state again after reconnection (but at least it will be in a somewhat understandable state).

(k) The operator interface and the higher level automation or monitoring fab computer system are removed by communication on the control network 120. These interfaces are provided from the cluster front end 6, so they do not create congestion on the control network 120 and do not affect the timing of communication with the secondary system 16.

(1) The data collection is completely isolated from a data network 140 that is different from the control communication on the control network 120. The separation of this function is described in more detail below.

Synchronous clock

The lithography system provides a clock signal to the control network to synchronize the motion within and across the subsystem. The system can provide two clock signals at different frequencies to provide clocks to sub-systems based on different timing accuracy, such as clock signals with accuracy of milliseconds and nanoseconds, respectively. For example, a pattern stream sub-system for streaming beam switching data and a beam switching subsystem for switching the electron beamlets to on or off need to exchange data at very high frequencies to enable beamlets. The high frequency switching, and the beam measurement subsystem requires the position measurement of the transmitted beam at a very high frequency.

Data network

Secondary system 16 is coupled to data network hub 14 via data network 140. Data collection, storage and management are performed via data network 140. Control network 120 forms a control network path between component control unit 12 and lithography etch subsystem 16, and data network 140 forms a data network between data hub 14 and lithography subsystem 16 path. Control network 120 and data network 140 are physically distinct networks. Each network has its own individual physical medium, including wiring, network components such as switches, and network connections to the secondary system 16. Thus, the control network path and the data network path contain physically distinct media and form different and independent communication paths.

Each lithography etch sub-system 16 has a connection to the control network 140 that is configured to receive and transmit control information to and from the component control unit 12 via the control network. Each lithography etch sub-system 16 has a separate connection to the data network 140 that is tuned to transmit data information to the data network hub 14 over the data network.

The data hub 14 is tuned to continuously record data received from the secondary system 16. Such information may include the measurement data adopted by the subsystem during the execution of the PJ, the setting of the secondary system, and the information used for debugging and fault tracing. Preferably, the data hub 14 always records all data (including low-level tracking data), so there is no need to initiate data logging. This continuous data record speeds up problem diagnosis, reduces the need for restarts, and reproduces errors or problems.

The component control unit 12 is coupled to the data hub 14 to communicate information about the progress of the PJs executed in the component control unit 12 for the data network The hub 14 records it. The data hub 14 continuously records the data received from the component control unit 12.

The data hub 14 contains a very large amount of data storage capacity sufficient to store an extremely large amount of low-level data from all of the secondary systems 16 over a long period of operation, such as a period of several months or years. The data stored by the data hub 14 is organized and tagged, preferably with a time stamp and a PJ identification. The stored data can be retrieved from the data hub 14 and analyzed and filtered. The preferred embodiment is performed offline.

The separation of the data collection through the data network 140 from the control communication through the control network 120 enables it to perform high frequency collection of large amounts of data on the data network without confusing communication over the control network. By controlling congestion on the control network 120 and avoiding large amounts of traffic appearing on the data network 140, the control network can operate in a quasi-instantaneous manner. The data management and storage functions in the data hub 14 are separated from the PJ execution in the component control unit 12 so that it can perform a large amount of data processing of the data network hub 14 without dragging the PJ processing of the component control unit 12. The separation of control and data collection and management makes the design simpler and does not require consideration of the interaction between the two systems.

The invention has been described above by way of specific embodiments. It is to be understood that the embodiments are susceptible to various modifications and alternative forms, which are well known to those skilled in the art and without departing from the spirit and scope of the invention. Accordingly, the scope of the invention is defined by the scope of the appended claims.

1‧‧‧Photolithography etching system

2‧‧‧Master Control System

3‧‧‧Group interface

4‧‧‧Operator console

5‧‧‧Communication component interface

6‧‧‧Photolithography etch cluster front end

7‧‧‧Photolithography etching system interface

10‧‧‧ lithography etching elements

12‧‧‧Component Control Unit (SUSC)

14‧‧‧Data Network Hub (SUSD)

15‧‧‧Plugin User (PIC)

16‧‧‧Photolithography etching system

18‧‧‧Pattern Data Processing System

19‧‧‧Actuator/pattern streamer (system)

20‧‧‧data path

22‧‧‧Substrate delivery system

23‧‧‧Service area

24‧‧‧Substrate supply system

100‧‧‧Photolithography etching system clustering

101-103, 105‧‧‧ communication

106‧‧‧ links

110‧‧‧Electronic source

113‧‧‧ collimator lens system

114a‧‧‧Pore array

114b‧‧‧Second pore array

116‧‧‧ concentrator lens array

117‧‧‧Small beam blanker array

118‧‧‧beam beam array

119‧‧‧beam deflector array

120‧‧‧Control Network/Projection Lens Array

121‧‧‧Substrate

130‧‧‧Expanded electron beam

133‧‧‧Small beam

134‧‧‧Subbeam

140‧‧‧Information Network

The description of the embodiments of the present invention is made with reference to the accompanying schematic drawings, and is merely exemplary. FIG. 1 is a schematic diagram of one embodiment of a network architecture for a lithography etching system according to the present invention; 2 is a schematic diagram of one embodiment of a network architecture for a lithography etching system including a group of lithographic etched elements; FIG. 3 is a schematic diagram of a layout of a lithographic etched element cluster; A simplified schematic diagram of one of the electro-optical cavities of a charged particle lithography etch element.

1‧‧‧Photolithography etching system

2‧‧‧Master Control System

3‧‧‧Group interface

4‧‧‧Operator console

5‧‧‧Communication component interface

6‧‧‧Photolithography etch cluster front end

7‧‧‧Photolithography etching system interface

10‧‧‧ lithography etching elements

12‧‧‧Component Control Unit (SUSC)

14‧‧‧Data Network Hub (SUSD)

15‧‧‧Plugin User (PIC)

16‧‧‧Photolithography etching system

18‧‧‧Pattern Data Processing System

19‧‧‧ pattern streamer

20‧‧‧data path

101-103, 105‧‧‧ communication

106‧‧‧ links

120‧‧‧Control network

140‧‧‧Information Network

Claims (16)

A cluster substrate processing system comprising a plurality of lithography etch elements (10), each lithography etch element being configured for independent exposure of a substrate according to pattern data, and each lithography etch element comprises: a plurality a lithography etching subsystem (16); a control network (120) configured to communicate control information between the lithography subsystem and the at least one component control unit (12), the component control The unit is configured to transmit instructions to the lithography etch subsystems and the lithography etch subsystem is configured to transmit a response to the component control unit; and a data network (140) configured to be used from the The lithography etching subsystem communicates with the data recording information of the at least one data network hub (14), the lithography etching subsystem is configured to transmit the data recording information to the data network hub, and the data network hub Configuring to receive and store the data record information, the cluster substrate processing system further includes a group of front ends (6) as an interface to an operator or a master system (2, 4), the group Poly front end is configured Transmitting control information to the at least one component control unit to control operation of exposure of the lithography subsystem to one or more wafers, and the cluster front end is further configured to receive the data network At least a portion of the data record information received by the hub, and the cluster substrate processing system further includes a data path (20) configured to use the pattern data to one or more of the lithography subsystems Transmission of a lithography etching subsystem, the data path forming a transmission path different from the control network and the data network, the pattern data does not need to pass the control The network or the data network is transmitted and transmitted to the one or more lithography etching subsystems, wherein the data path includes a pattern stream system (19). The cluster substrate processing system of claim 1, wherein the control network forms a control network path between the component control unit and the lithography etching subsystem, and the data network is A data network hub forms a data network path with the lithography etch subsystems, and wherein the control network path and the data network path comprise physically distinct media. A grouping substrate processing system according to claim 1 or 2, wherein each lithography etching system has a first connection to the control network, the first connection being configured to pass through the control network Receiving an instruction from the component control unit and transmitting a response to the component control unit, and each lithography etching system has a different second connection to the data network, the second connection being modulated to transmit the data The network transmits the data record information to the data network hub. The cluster substrate processing system of claim 1 or 2, wherein the component control unit is configured to transmit an instruction to the complex lithography etching subsystem through the control network and receive from a lithography The relevant information of the response of the etching subsystem is communicated to the data network hub. A grouping substrate processing system according to claim 1 or 2, wherein during operation of one of the lithographic etch elements, a lithography etching subsystem in the lithography etch elements is configured to transmit The data network continuously transmits the recorded data to the data network hub. Grouped substrate processing system according to claim 1 or 2 The data hub stores information required to initialize the lithography subsystem, and initialization information for a lithography subsystem is transmitted to the lithography during the beginning of the lithography system Etching the secondary system. A grouping substrate processing system according to claim 1 or 2, wherein at least one lithography etching subsystem of the lithography subsystem is configured to transmit information to the lithography etching system at the beginning A component control unit that indicates the presence of one of the lithography subsystems on the control network and one of the lithography subsystems identifies an identity and indicates one or more instructions that the lithography subsystem can execute. A cluster substrate processing system according to claim 1 or 2, wherein the control network is configured for communication of service information between the component control unit and the lithography etching subsystem, and the group The poly front end is tuned to provide service information for performing service functions on one or more of the lithographic etch elements. A cluster substrate processing system according to claim 1 or 2, wherein the system is configured to provide a second clock signal to the lithography subsystem, one of the two clock signals having a higher than the other A clock frequency having a clock frequency of at least 100 times, and wherein the lithography subsystems are configured to synchronize their actions using the clock signals. A grouping substrate processing system according to claim 1 or 2, wherein the lithography etching subsystem does not provide any interface to an operator or a master system, but provides the communication only through the cluster front end To the interface of the lithography etching subsystem. A method for data communication in a cluster substrate processing system The grouped substrate processing system includes a plurality of lithography etch elements (10), each lithographic etch element being configured for independent exposure of the substrate in accordance with pattern data, and each lithographic etch element comprises a control network (120), connected to a component control unit (12) and a plurality of lithography etching subsystems (16), a data network (140), connected to a data network hub (14) and the plurality of micro a shadow etching subsystem (16), and a group of poly front ends (6), the method comprising: transmitting control information from the component control unit (12) to the lithography etching subsystem (16) through the control network (120) One or more lithography etching subsystems; transmitting a response from the lithography subsystem (16) to the component control unit (12) through the control network (120); through the data network (140) Receiving data record information from the lithography subsystem (16) to the data network hub (14); receiving and storing the data record information in the data network hub (14); from the cluster front end (6) Transmitting control information to the component control unit (12) to control the lithography etching subsystem for one or The operation of exposing the substrates; transmitting at least a portion of the data record information stored by the data hub (14) to the cluster front end (6); and transmitting the pattern data through a data path (20) to the The one or more lithography etching subsystems in the lithography subsystem, the data path forming a transmission path different from the control network and the data network, and the pattern data does not need to pass through the control network Or the data network is transmitted and transmitted to the one or more lithography etching subsystems. According to the method of claim 11, wherein the control information transmitted to the lithography subsystem and the response from the lithography subsystem are transmitted to the component control unit only through the control network, and from the The data recording information of the lithography etching system is transmitted to the data network hub only through the data network. According to the method of claim 11 or 12, further comprising transmitting, from the component control unit, the instruction relating to the transmission to the complex lithography etching system and the data received from a lithography etching system to the data network Road hub. The method of claim 11 or 12, wherein transmitting the data record information from a lithography etch subsystem to the data hub is performed continuously during execution of the lithography subsystem. According to the method of claim 11 or 12, the information required to initialize the lithography etching subsystem is stored in the data network hub and transmitted during the beginning of a lithography etching system. The initialization information is sent to the lithography etching subsystem. According to the method of claim 11 or 12, further comprising transmitting information from the lithography etching system to the component control unit at the beginning of a lithography etching system in the lithography etching subsystem, the information indication The presence of the lithography etch subsystem on the control network and one of the lithography subsystems identifies the identity and indicates one or more instructions that the lithography subsystem can execute.